Non-transitory computer-readable storage medium for storing prediction program, prediction method, and prediction apparatus

Abstract

A non-transitory computer-readable storage medium storing a prediction program for causing a computer to perform processing including: listing combinations of feature amounts that are correlated with a target label; creating a policy to achieve the target label for a prediction target based on a difference between the listed combinations of the feature amounts and a combination of feature amounts of the prediction target; and determining appropriateness of the created policy based on performance information that indicates past performances.

What is claimed is: 
     
         1 . A non-transitory computer-readable storage medium storing a prediction program for causing a computer to perform processing including:
 listing combinations of feature amounts that are correlated with a target label;   creating a policy to achieve the target label for a prediction target based on a difference between the listed combinations of the feature amounts and a combination of feature amounts of the prediction target; and   determining appropriateness of the created policy based on performance information that indicates past performances.   
     
     
         2 . The non-transitory computer-readable storage medium according to  claim 1 , wherein,
 in the processing of creating, a combination of feature amounts that corresponds to the difference is created as the policy.   
     
     
         3 . The non-transitory computer-readable storage medium according to  claim 2 , wherein,
 in the processing of determining, it is determined that the policy is appropriate in a case where an event related to the combination of the feature amounts included in the policy occurs at a predetermined occurrence frequency in the past performances included in the performance information.   
     
     
         4 . The non-transitory computer-readable storage medium according to  claim 3 , wherein,
 in the processing of listing, the combinations of the feature amounts are re-listed by setting, as uncontrollable, the feature amounts included in the combinations in a case where the event related to the combination of the feature amounts included in the policy does not occur at the predetermined occurrence frequency in the past performances included in the performance information, and   in the processing of creating, the policy is re-created based on a difference between the re-listed combinations of the feature amounts and the combination of the feature amounts of the prediction target.   
     
     
         5 . The non-transitory computer-readable storage medium according to  claim 1 , wherein,
 in the processing of creating, a plurality of the policies is created with a probability of achieving the target label, and   a computer is further caused to execute processing of outputting a policy determined to be appropriate among the plurality of created policies together with the probability.   
     
     
         6 . The non-transitory computer-readable storage medium according to  claim 1 , wherein
 the label is a result associated with a predetermined event of the prediction target or a target different from the prediction target, and   the performance information includes at least performance information regarding the predetermined event of the prediction target or the target different from the prediction target.   
     
     
         7 . A prediction method implemented by a computer, the prediction method comprising:
 listing combinations of feature amounts that are correlated with a target label;   creating a policy to achieve the target label for a prediction target based on a difference between the listed combinations of the feature amounts and a combination of feature amounts of the prediction target; and   determining appropriateness of the created policy based on performance information that indicates past performances.   
     
     
         8 . The prediction method according to  claim 7 , wherein,
 in the processing of creating, a combination of feature amounts that corresponds to the difference is created as the policy.   
     
     
         9 . The prediction method according to  claim 8 , wherein,
 in the processing of determining, it is determined that the policy is appropriate in a case where an event related to the combination of the feature amounts included in the policy occurs at a predetermined occurrence frequency in the past performances included in the performance information.   
     
     
         10 . The prediction method according to  claim 9 , wherein,
 in the processing of listing, the combinations of the feature amounts are re-listed by setting, as uncontrollable, the feature amounts included in the combinations in a case where the event related to the combination of the feature amounts included in the policy does not occur at the predetermined occurrence frequency in the past performances included in the performance information, and   in the processing of creating, the policy is re-created based on a difference between the re-listed combinations of the feature amounts and the combination of the feature amounts of the prediction target.   
     
     
         11 . The prediction method according to  claim 7 , wherein,
 in the processing of creating, a plurality of the policies is created with a probability of achieving the target label, and   a computer is further caused to execute processing of outputting a policy determined to be appropriate among the plurality of created policies together with the probability.   
     
     
         12 . The prediction method according to  claim 7 , wherein
 the label is a result associated with a predetermined event of the prediction target or a target different from the prediction target, and   the performance information includes at least performance information regarding the predetermined event of the prediction target or the target different from the prediction target.   
     
     
         13 . A prediction apparatus comprising a control unit that executes processing including:
 listing combinations of feature amounts that are correlated with a target label;   creating a policy to achieve the target label for a prediction target based on a difference between the listed combinations of the feature amounts and a combination of feature amounts of the prediction target; and   determining appropriateness of the created policy based on performance information that indicates past performances.   
     
     
         14 . The prediction apparatus according to  claim 13 , wherein,
 in the processing of creating, a combination of feature amounts that corresponds to the difference is created as the policy.   
     
     
         15 . The prediction apparatus according to  claim 14 , wherein,
 in the processing of determining, it is determined that the policy is appropriate in a case where an event related to the combination of the feature amounts included in the policy occurs at a predetermined occurrence frequency in the past performances included in the performance information.   
     
     
         16 . The prediction apparatus according to  claim 15 , wherein,
 in the processing of listing, the combinations of the feature amounts are re-listed by setting, as uncontrollable, the feature amounts included in the combinations in a case where the event related to the combination of the feature amounts included in the policy does not occur at the predetermined occurrence frequency in the past performances included in the performance information, and   in the processing of creating, the policy is re-created based on a difference between the re-listed combinations of the feature amounts and the combination of the feature amounts of the prediction target.   
     
     
         17 . The prediction apparatus according to  claim 13 , wherein,
 in the processing of creating, a plurality of the policies is created with a probability of achieving the target label, and   a computer is further caused to execute processing of outputting a policy determined to be appropriate among the plurality of created policies together with the probability.   
     
     
         18 . The prediction apparatus according to  claim 13 , wherein
 the label is a result associated with a predetermined event of the prediction target or a target different from the prediction target, and the performance information includes at least performance information regarding the predetermined event of the prediction target or the target different from the prediction target.